Molecular Characterization and Antimicrobial Susceptibilities of Nocardia Species Isolated from the Soil; a Comparison with Species Isolated from Humans

Home , Nocardia, Nocardia brasiliensis, Nocardia farcinica

microorganisms

Article Molecular Characterization and Antimicrobial Susceptibilities of Nocardia Species Isolated from the Soil; A Comparison with Species Isolated from Humans

Gema Carrasco 1, Sara Monzón 2, María San Segundo 1, Enrique García 3, Noelia Garrido 1, María J. Medina-Pascual 1, Pilar Villalón 1 , Ana Ramírez 3, Pilar Jiménez 4, Isabel Cuesta 2 and Sylvia Valdezate 1,*

1 Reference and Research Laboratory for Taxonomy, National Centre of Microbiology, Instituto de Salud Carlos III, Majadahonda, 28220 Madrid, Spain; [email protected] (G.C.); [email protected] (M.S.S.); [email protected] (N.G.); [email protected] (M.J.M.-P.); [email protected] (P.V.) 2 Bionformatics Unit, Applied Services, Training and Research, Instituto de Salud Carlos III, Majadahonda, 28220 Madrid, Spain; [email protected] (S.M.); [email protected] (I.C.) 3 Soil Microbiology Laboratory, Microbiology and Parasitology Department, Pharmacy and Bioanalys Faculty, Los Andes University, Mérida 5101, Venezuela; [email protected] (E.G.); [email protected] (A.R.) 4 Genomics Laboratory, Applied Services, Training and Research, Instituto de Salud Carlos III, Majadahonda, 28220 Madrid, Spain; [email protected] * Correspondence: [email protected]; Tel.: +34-91-822-3734; Fax: +34-91-509-7966

Received: 22 May 2020; Accepted: 11 June 2020; Published: 15 June 2020

Abstract: Nocardia species, one of the most predominant Actinobacteria of the soil microbiota, cause infection in humans following traumatic inoculation or inhalation. The identification, typing, phylogenetic relationship and antimicrobial susceptibilities of 38 soil Nocardia strains from Lara State, Venezuela, were studied by 16S rRNA and gyrB (subunit B of topoisomerase II) genes, multilocus sequence analysis (MLSA), whole-genome sequencing (WGS), and microdilution. The results were compared with those for human strains. Just seven Nocardia species with one or two strains each, except for Nocardia cyriacigeorgica with 29, were identified. MLSA confirmed the species assignments made by 16S rRNA and gyrB analyses (89.5% and 71.0% respectively), and grouped each soil strain with its corresponding reference and clinical strains, except for 19 N. cyriacigeorgica strains found at five locations which grouped into a soil-only cluster. The soil strains of N. cyriacigeorgica showed fewer gyrB haplotypes than the examined human strains (13 vs. 17) but did show a larger number of gyrB SNPs (212 vs. 77). Their susceptibilities to antimicrobials were similar except for beta-lactams, fluoroquinolones, minocycline, and clarithromycin, with the soil strains more susceptible to the first three (p 0.05). WGS was performed on four strains belonging to the soil-only cluster and ≤ on two outside it, and the results compared with public N. cyriacigeorgica genomes. The average nucleotide/amino acid identity, in silico genome-to-genome hybridization similarity, and the difference in the genomic GC content, suggest that some strains of the soil-only cluster may belong to a novel subspecies or even a new species (proposed name Nocardia venezuelensis).

Keywords: Nocardia spp.; Nocardia cyriacigeorgica; soil; nocardiosis; gyrB; MLSA; WGS; antimicrobial resistance; potential new species/subspecies N. venezuelensis

Microorganisms 2020, 8, 900; doi:10.3390/microorganisms8060900 www.mdpi.com/journal/microorganisms Microorganisms 2020, 8, 900 2 of 22

Importance This study highlights hitherto unreported identification, typing, phylogenetic relationships, and antimicrobial susceptibility among soil Nocardia strains, with special reference to N. cyriacigeorgica, the main species detected and one of the most common causes of human nocardiosis. The work also compares soil strains carrying seven Nocardia species with clinical strains isolated from humans, detecting only genetic and antimicrobial susceptibilities differences in N. cyriacigeorgica. Our findings suggest that some of the soil N. cyriacigeorgica strains detected may, in fact, belong to a new subspecies or even a new species (proposed name Nocardia venezuelensis).

1. Introduction Nocardia spp. species are found everywhere from sludge and soil to contaminated soil water, deep-sea sediments [1], and desert habitats [2]. Some even infect plants and animals [3–5]. They are among the most predominant Actinobacteria of the soil microbiota, including that of the extreme biosphere [6]. The members of Nocardia spp. are producers of diverse natural bioactive metabolites [7], such as antimicrobials, enzyme inhibitors, immunomodifiers, and plant growth-promoting substances, etc. [8,9], a result of the physiological and biochemical pressures imposed by the environmental conditions under which they live [10]. Their activity in the degradation of polycyclic aromatic hydrocarbon [11,12] focused on them as potential xenobiotic bioremediators. Although they cause a number of severe invasive diseases [3,13], the members of Nocardia spp. are mainly opportunistic pathogens in humans, usually affecting the lungs, central nervous system, and skin [14,15]. The burden of human nocardiosis differs between geographical locations. In previous work, Nocardia strains were isolated from soil collected at different sites in Lara State (Venezuela) [16], where the prevalence of human mycetoma (a severe cutaneous infection) is high. The present work examines the identity of these strains via 16S rRNA and gyrB genes analysis, together with multi-locus sequence analysis (MLSA), whole-genome sequencing (WGS), and susceptibilities. With a special focus on the most prevalent soil species detected, differences and similarities with clinical strains were explored.

2. Materials and Methods

2.1. Molecular Identification of Species In the present work, 38 phenotypically identified strains were submitted to the National Centre for Microbiology (CNM, Majadahonda, Madrid, Spain) for molecular identification. The strains were isolated from soil samples collected over two periods-08/2002 and 05/2006-from nine sites in six municipalities in Lara State, NW Venezuela (Figure1). Table1 shows the climatic characteristics of each site, and the Supplemental File the soil culture and the phenotypic identification previously described [16]. After growth on Columbia agar supplemented with 5% (v/v) sheep0s blood and buffered charcoal–yeast extract agar (BCYE) for 48–72 h at 37 ◦C under aerobic conditions, their chromosomal DNA was extracted by the boiling method. The 16S rRNA and gyrB genes were then amplified and sequenced as previously described [17], and species identified by comparing them against type strain sequences [18,19] using the BLAST algorithm v.2.2.10 (http://www.ncbi.nlm.nih.gov/BLAST). Similarity values of 99.6% for 16S rRNA [20], and 93.5% for gyrB, were deemed to indicate the ≥ ≥ same species [19]. Sequences were assembled using SEQ-Man software (DNASTAR, Inc., Madison, WI) and, using BioEdit [21], adjusted for phylogenetic analysis to coincide with the length of the shortest sequence for each reference strain (16S rRNA 1215 bp; gyrB 726 bp). They were then aligned using the ClustalW algorithm [22], and 16S rRNA and the gyrB phylogenetic trees constructed using MEGA 6 software [23] following the neighbor-joining (NJ) and maximum likelihood (ML) methods [24] with 1000 bootstrap replications. Microorganisms 2020, 8, 900 3 of 22

Table 1. Characteristics of the Nocardia spp. soil strains isolated in Lara State, Venezuela.

Percentage Identity Percentage Identity Drug Latitude Species with Respect to with Respect to Location Temperature Sample Time Strain No. Resistance (N)/Longitude(W) Weather Type (16S rDNA) DMS 44484T DMS 44484T (Municipality) /Rainfall (m, yr) Phenotype and Altitude 16S rDNA gyrB 20110625 N. cyriacigeorgica 99.84% 99.4% CIP CLA Arenales (Torres) 10 9 11" Semi-arid ◦ 0 27 C/400mm August, 2002 69 54´12" ◦ continental 20110626 b,c,d N. cyriacigeorgica 100% 95.9% xl CIP CLA Arenales ◦ 517m 20110630 N. cyriacigeorgica 99.84% 93.1% Xl CLA Caraquita(Crespo) 20110631 N. cyriacigeorgica 99.84% 93.1% Xl CLA Caraquita 20110632 N. cyriacigeorgica 99.84% 93.0% XL cla Caraquita 20110634 N. cyriacigeorgica 99.84% 93.1% XL CLA Caraquita 10◦ 410 11" Subhumid 20110637 N. cyriacigeorgica 99.84% 93.1% CLA Caraquita 69◦ 050 11" 25 ◦C/743mm April, 2006 interior 20110638 N. cyriacigeorgica 99.84% 92.8% xl Caraquita 685m (transitional) 20110640 N. cyriacigeorgica 98.85% 93.1% CLA Caraquita 20110651 a,c N. cyriacigeorgica 99.84% 86.2% XL CLA CIP Caraquita 20110652 a,c N. vermiculata - - - Caraquita

20110643 a,c N. vermiculata - - CIP El Padrón (Torres) 10 20 44” Subhumid ◦ 0 May, 2006 70 28 59 ” 26 ◦C/921mm continental a,b,c N. elegans - - XL TOB CIP El Padrón ◦ 0 0 20110644 643m Subhumid 20110624 d N. cyriacigeorgica 99.92% 95.9% CLA Humocaro (Morán) 9o 40´57" August, 2002 continental 69o 58 12" 24 ◦C/700mm 20110658 cyriacigeorgica 99.84% 87.9% Xl CLA Humocaro (seasonal) 964m o Semi-arid 20110622 N. cyriacigeorgica 99.84% 93.3% CLA CIP Potrero de Bucare 10 18´51" 24 ◦C/700mm (Iribarren) 69o 27´45" 25 C/339mm continental 20110623 N. cyriacigeorgica 99.75% 98.2% CIP◦ August, 2002 711m Quebrada de Oro 20110627 a,c N. abcessus - - IMI CIP (Crespo) 20110628 N. abcessus - - IMI CIP Quebrada de Oro 20110633 N. cyriacigeorgica 99.84% 93.1% CLA Quebrada de Oro 20110635 a,c N. asteroides - - - Quebrada de Oro 20110636 N. cyriacigeorgica 99.92% 93.1% CIP cla Quebrada de Oro d 20110639 N. cyriacigeorgica 99.92% 92.6% CIP CLA Quebrada de Oro 10◦ 160 2" Subhumid 20110641 N. cyriacigeorgica 99.84% 93.1% CLA Quebrada de Oro 69◦ 20 22" 24 ◦C/1285mm April, 2006 interior 20110642 a,c N. cyriacigeorgica 99.84% 85.1% xl Quebrada de Oro 1278m (transitional) 20110645 N. cyriacigeorgica 99.84% 93.1% XL CLA min Quebrada de Oro 20110646 N. cyriacigeorgica 99.84% 93.1% XL CLA CIP Quebrada de Oro 20110647 N. cyriacigeorgica 100% 93.1% CLA Quebrada de Oro 20110648 d N. cyriacigeorgica 99.84% 92.8% CLA Quebrada de Oro 20110649 d N. cyriacigeorgica 99.84% 91.2% CLA Quebrada de Oro 20110650 N. cyriacigeorgica 99.92% 93.1% XL, CLA Quebrada de Oro Microorganisms 2020, 8, 900 4 of 22

Table 1. Cont.

9◦ 470 2” Subhumid d 20110629 N. cyriacigeorgica 99.75% 92.4% CLA Sarare (Simón Planas) 69◦ 90 40” 26 ◦C/1434mm August, 2002 continental 269m (seasonal) 20110616 b,c N. cyriacigeorgica 100% 85.0% CIP, CLA Siquisique (Urdancia) 20110617 a,c N. rhamnosiphila - - CLA Siquisique 10o 34´24" 20110618 a,c N. rhamnosiphila - - - Siquisique Semi-arid 69o 42´ 5" 27 C/358mm August, 2002 20110619 N. cyriacigeorgica 99.92% 99.4% XL CLA CIP Siquisique ◦ continental 271m 20110620 N. cyriacigeorgica 100% 99.4% XL CLA Siquisique Xl IMI tob 20110621 a,b N. mexicana -- Siquisique CLA min The vegetation at all sites was thorny scrub, except for the Caraquita, Quebrada de Oro, and El Padrón site, which was forested. The minimum inhibitory concentrations (MIC) values were categorized following the Clinical Laboratory Standard Institute interpretative criteria (CLSI, 2018). Resistant and intermediate values are coded in capital and lowercase respectively. Antimicrobial acronyms: amoxicillin/clavulanate (XL), tobramycin (TOB), clarithromycin (CLA), minocycline (MIN), ciprofloxacin (CIP), trimethoprim/sulfamethoxazole (SxT). a Disagreement in identification between the 16S rRNA and gyrB techniques; b disagreement in identification between the 16S rRNA and multilocus sequence analysis (MLSA) techniques; c disagreement in identification between the gyrB and MLSA techniques; d strains studied by whole-genome sequencing (WGS). Microorganisms 2020, 8, 900 5 of 22 Microorganisms 2020, 8, x FOR PEER REVIEW 3 of 22

Figure 1. GeographicGeographic distribution distribution of soil soil NocardiaNocardia strainsstrains collected collected in in Lara Lara State State (Venezuela). (Venezuela). Numbers indicate the number of strains isolated per site.

2.2. Multilocus Sequence Analysis (MLSA) All 38 soil strains were then subjected to MLSA [25] alongside a further ﬁve Venezuelan strains (three of N. cyriacigeorgica and two of N. farcinica), eight Spanish clinical Nocardia strains, and type strain sequences retrieved from GenBank. MLSA was performed using trimmed sequences of concatenated gyrB-16S rRNA-secA1-hsp65 (1790-bp). A NJ phylogenetic tree was then constructed using MEGA 6 software [23]. It should be noted that N. elegans lacks a reference type strain for all the genes here examined; the clinical N. elegans 20130578 strain was used as an alternative, and it is, therefore this strain that appears in the phylogenetic tree.

2.3. Genetic Similarities Among Soil and Clinical Nocardia cyriacigeorgica Strains Given the strong predominance of Nocardia cyriacigeorgica strains in the sampled soils, 30 previously characterized Spanish clinical strains belonging to this species [17] were compared with them in terms of their 16S rRNA, gyrB, and GyrB (DNA gyrase subunit B) sequences. Hunter-Gaston discrimination indices (HGDI) [26], single nucleotide polymorphisms (SNPs), and haplotype numbers were examined using DnaSP software [22]. The N. cyriacigeorgica 16S rRNA, gyrB, and GyrB sequences of the type strain DSM 44484T (GenBank accession number AF430027, GQ496121, and ACV89678, respectively) were used to determine SNP numbers. In addition, the population structures of the soil and clinical groups were examined via a gyrB NJ phylogenetic tree, with the inclusion of a further 3 Venezuelan clinical strains and the genome reference strain GUH-2 (GenBank accession number FO082843). Microorganisms 2020, 8, 900 6 of 22

2.4. Antimicrobial Susceptibilities The antimicrobial susceptibilities of all the strains were determined according to CLSI M24-A2 guidelines, using the corresponding control strains [20] and employing the microdilution method with RAPMYCO panels (ThermoFisher, Inc., Cleveland, OH, USA). These panels contain amikacin (AMI), amoxicillin/clavulanic acid (AUG2), cefepime (FEP), cefoxitin (FOX), ceftriaxone (AXO), ciprofloxacin (CIP), clarithromycin (CLA), doxycycline (DOX), imipenem (IMI), linezolid (LZD), minocycline (MIN), moxifloxacin (MXF), tigecycline (TGC), tobramycin (TOB), and co-trimoxazole (trimethoprim/sulfamethoxazole, SXT). Minimum inhibitory concentrations (MIC) were determined following Clinical Laboratory Standard Institute interpretative criteria [27]; intermediate values were categorized as resistant. Susceptibility to trimethoprim/sulfamethoxazole and linezolid was tested using the E-test (bioMérieux, Marcy-l’Étoile, France). Susceptibility rates across strains belonging to the main species from the soil and human sources were compared using the χ2 test or two-tailed Fisher’s exact test as required. Significance was set at p 0.05. All calculations were performed using ≤ STATA v.13.1 software (StataCorp, College Station, TX, USA).

2.5. Bioinformatic Analysis Six representative soil N. cyriacigeorgica strains, thought to be distinct according to their gyrB analysis results, were sequenced. Genomic DNA was extracted from single subcultured colonies using the QIAamp DNA Mini Kit (Qiagen, Hilden, Germany). Paired-end libraries were prepared using the Nextera-XT DNA Library Preparation Kit (Illumina 1.9, San Diego, CA, USA) and sequencing performed using the Illumina NextSeq 500 platform (mean sequencing depth 90 per sample). Read quality control was ∼ × undertaken using FastQC v. 0.11.8 software. Trimmomatic v. 0.33 software [28] was used to remove adapter contamination and to trim low-quality regions (phred >2 0 in a 4 nt window, minimum length 70 bp). Kmerfinder v. 3.0 software [29] was then used for species confirmation and to detect contamination. Assembly was performed using Spades v. 3.8.0 software [30]; Prokka v. 1.12 software [31] was used for genome annotation. Quast v. 4.1 software [32] was used for assembly quality control. Species assignations were confirmed by comparing the average nucleotide identity (ANI) (https://www.ezbiocloud.net/tools/ani)[33], average amino acid identity (AAI) (http://enveomics.ce.gatech.edu/aai)[34], and in silico genome-to-genome distance similarity (GGDH; DDH-estimate) (https://ggdc.dsmz.de/ggdc.php#)[35] results against the N. cyriacigeorgica GUH-2 (NC_016887.1) reference genome, the genome of the type strain DSM 44484T (NZ_VBUR00000000.1), and other genomes [36]. The AAI-profiler (http://ekhidna2.biocenter.helsinki.fi/AAI)[37], TrueBacTM IDBETA (https://www.truebacid.com/genome/)[38], and the Type Strain Genome Server (TYGS) (https://tygs.dsmz.de)[39] web servers were also used to resolve taxonomic identities. High-quality assemblies of the same six soil Nocardia strains, the type strain DSM 44484T, the reference genome of N. cyriacigeorgica GUH-2, and five genome assemblies for N. cyriacigeorgica (available in NCBI at the time of publication) were subjected to core-genome gene-by-gene typing (cgMLST) using chewBBACA v. 2.0.17.2 software (open-source in https://github.com/B-UMMI/ chewBBACA)[40]. Those loci corresponding to potentially complete coding sequences (CDS) that were unique, but present in 95% of the strains, were used in subsequent phylogenetic analysis, using GrapeTree v .2.0 software to visualize the results. A phylogenetic analysis was also performed using bcgTree v .1.1.0 software (available at https://github.com/iimog/bcgTree)[41]; this searches for 107 conserved proteins among the examined bacteria and creates a concatenated gene matrix for a maximum likelihood phylogeny analysis with 100 bootstrap replications (performed using RAxML v .8.2.9 software) [42]. Microorganisms 2020, 8, 900 7 of 22

Antimicrobial resistance genes were searched using different tools as (date last accessed, May 2020): ResFinder (identification threshold of 90%) [43], Antibiotic-Resistant Target Seeker (ARTS) [44], the Comprehensive Antibiotic Resistance Database (CARD, with strict criteria in RGI) [45] and by KOALA for KEGG Orthology [46]. Additionally, the SRST2 program [47], was used to detect resistance genes and alleles with the ARGannot database [48]. Phages were identified using PHASTER software (PHAge Search Tool Enhanced Release) (https://phaster.ca/)[49].

2.6. Accession Number(s) This Whole Genome Shotgun project has been deposited at DDBJ/ENA/GenBank under the accession numbers JAAGVC000000000, JAAGVB000000000, JAAGVA000000000, JAAGUZ000000000, JAAGUY000000000 and JAAGUX000000000. https://www.ncbi.nlm.nih.gov/genome/.

3. Results

3.1. Distribution of Nocardia Species in the Soil The number of Nocardia strains recovered from the soil samples ranged from 1–14 (mean = 6 strains per sample). The Quebrada de Oro (14 strains) and Caraquita (9 strains) sites returned the highest number of soil strains. 16S rRNA sequencing [50] identified the species of all 38 strains with the following distribution: N. cyriacigeorgica 29 strains, N. abscessus 2, N. rhamnosiphila 2, N. vermiculata 2, N. asteroides 1, N. elegans 1, and N. mexicana 1 strain. Three different species were found in Quebrada de Oro and Siquisique, in the Crespo and Urdancia municipalities, respectively. The most common species, N. cyriacigeorgica, was present at all sites except for El Padrón (in the Torres municipality) (Table1). Species assignment via gyrB analysis [19] agreed with the 16S rDNA-based identifications for 27 strains (71.05%). Table1 highlights those for which the results were discrepant.

3.2. Phylogenetic Analysis by MLSA MLSA assigned all the soil strains but four to the same species as determined by 16S rRNA analysis (Table1). The percentage similarity of each MLSA sequence with respect to the MLSA sequence of the respective type strain was: 94.0–98.3% in N. cyriacigeorgica, 93.2–94.5% in N. abscessus, 93.0–93.5% in N. rhamnosiphila, 96.3% in N. asteroides, and 93.5% in N. mexicana. In addition, MLSA conﬁrmed the gyrB-based identiﬁcation of 27 strains. The MLSA phylogenetic tree showed the 38 soil strains to group into three clusters for NJ (Figure2), and more for ML topologies (Supplementary Figure S1). Most gathered into cluster A, N. elegans, N. nova and N. vermiculata grouped into cluster C, and N. cyriacigeorgica strains were found in all three clusters. The clinical strains fell closer to the type strain of each species than did the soil strains. Twenty of the 29 soil N. cyriacigeorgica strains fell into cluster B: the remainder were distributed across clusters C (n = 7) and A (n = 2).

3.3. Antimicrobial Susceptibilities Table1 shows the antimicrobial-resistance phenotypes for each soil strain. These soil strains showed a phenotype that ﬁtted the drug pattern type [51], except for the N. asteroides soil strain which was susceptible to aminoglycosides and clarithromycin. The N. mexicana strain showed a wider resistance spectrum. The soil strains of N. cyriacigeorgica showed variable resistance to amoxicillin-clavulanate, clarithromycin, and ciproﬂoxacin. Table2 shows the corresponding MIC50, MIC90, MIC range, and resistance rates. Microorganisms 2020, 8, x FOR PEER REVIEW 8 of 22

Microorganismstype strain2020 of, 8 ,each 900 species than did the soil strains. Twenty of the 29 soil N. cyriacigeorgica strains fell8 of 22 into cluster B: the remainder were distributed across clusters C (n = 7) and A (n = 2).

Figure 2. Phylogenetic tree based on the MLSA neighbor-joining (NJ) analysis (gyrB-16S rRNA-secA-hsp65 genes) of the 38 N. cyriacigeorgica soil strains (in blue), six Venezuelan and nine Spanish clinical strains (in red), plus the type strains (in black). The asterisk indicates the strains selected for WGS. Na stands for N. abscessus, Nast for N. asteroides, Nc for N. cyriacigeorgica, Ne for N. elegans, Nf for N. farcinica, Ni for N. ignorata, Nm for N. mexicana, Nr for N. rhamnosiphila, and Nv for N. vermiculata. The reliability of the topologies was assessed by the bootstrap method (1000 replications). Microorganisms 2020, 8, 900 9 of 22

Table 2. Antimicrobial susceptibilities of 29 N. cyriacigeorgica soil strains and 30 N. cyriacigeorgica clinical strains. Comparison of resistance rates.

Resistance (%) Sign. Antimicrobial Agent MIC (mg/L) 1 2–4 Difference Range MIC50 MIC90 (p 0.05) ≤ Amoxicillin-clavulanic acid 4 Soil 2–32 8 32 14 (48.27%) ≤ yes Clinical 2–64 32 32 23 (76.7%) ≤ Cefoxitin Soil 4–128 8 32 12 (41.4%) 5 ≤ yes Clinical 4– 128 128 128 28 (93.4%) 5 ≤ ≥ ≥ Ceftriaxone Soil 4 4 4 0 ≤ ≤ ≤ yes Clinical 4–16 4 8 5 (16.7%) ≤ ≤ Cefepime Soil 1–16 2 8 1 (3.4%) ≤ yes Clinical 1–32 16 32 18 (60.0%) ≤ Imipenem Soil 2–4 2 2 0 ≤ ≤ ≤ yes Clinical 2–32 8 32 24 (79.2%) ≤ Amikacin Soil 1–16 1 2 1 (3.4%) ≤ ≤ no Clinical 1–16 1 1 1 (3.0%) ≤ ≤ ≤ Tobramycin Soil 1–2 1 1 0 ≤ ≤ ≤ no Clinical 1–16 1 1 2 (6.6%) ≤ ≤ ≤ Ciprofloxacin Soil 0.12– 4 1 4 8 (28.5.0%) ≤ ≥ ≥ yes Clinical 2– 4 4 4 29 (96.7%) ≥ ≥ ≥ Moxifloxacin Soil 0.25– 4 0.5 4 10 (34.5%) ≤ ≥ yes Clinical 1– 4 4 4 29 (96.7%) ≥ ≥ Clarithromycin Soil 1– 16 8 16 28 (96.5%) ≥ yes Clinical 0.06– 16 16 16 22 (73.3%) ≤ ≥ ≥ ≥ Doxycycline Soil 0.25–8 2 4 17 (58.6%) ≤ no Clinical 0.12–8 2 4 19 (62.7%) ≤ Minocycline Soil 1–8 1 2 7 (24.1%) ≤ ≤ yes Clinical 1–4 2 4 18 (59.4%) ≤ Tigecycline Soil 0.06–4 0.25 1 -6 no Clinical 0.25– 4 0.25 2 -6 ≤ ≥ Co-trimoxazole 4 Soil 0.25–0.5 0.25 0.5 0 ≤ no Clinical 0.25–4 0.5 2 1 (3.3%) ≤ Linezolid Soil 1–2 1 1 0 ≤ ≤ ≤ no Clinical 1–4 1 2 0 ≤ ≤ 1 Minimum inhibitory concentrations, MIC50 and MIC90 are the MICs at which 50% and 90% of the strains were inhibited respectively. 2 Clinical Laboratory Standard Institute intermediate and resistant criteria (document M24-A2) (values expressed in mg/L): amoxicillin/clavulanate, cefepime and cefoxitin (XL, FEP, FOX, 16, 32); ceftriaxone (AXO, 16–32, 64); imipenem (IMI, 8, 16); amikacin (AMI, -, 16); tobramycin (TOB, 8, 16); ciprofloxacin≥ and moxifloxacin(CIP≥ and MXF, 2, 4); clarithromycin≥ (CLA, 4, 8);≥ doxycycline and minocycline≥ (DOX and MIN, 2–4, 8); trimethoprim/sulfamethoxazole≥ (SXT, -, 4/76); linezolid≥ (LZD, 16); 3 Number and percentage of intermediate and≥ resistant strains; 4 Concentrations of amoxicillin≥ /clavulanate (ratio≥ 2:1) and trimethoprim/sulfamethoxazole (ratio 1:19) are expressed in terms of amoxicillin and trimethoprim respectively; 5 The available breakpoint for cephalosporins was used ( 8 mg/L); 6 No available breakpoint. ≥ Microorganisms 2020, 8, 900 10 of 22

3.4. Comparison of Soil and Human N. cyriacigeorgica Strains Table S1 and Table2 compare the 16S rRNA, gyrB, and GyrB sequences and antimicrobial susceptibilities of the soil N. cyriacigeorgica strains to those reported for previously studied Spanish human strains [17]. The soil strains were represented by three 16S haplotypes while the human strains were represented by just one, and by 13 gyrB haplotypes rather than 17 for the human strains. The high HGDI of the clinical N. cyriacigeorgica strains showed them to be more diverse than the soil strains (0.94 vs. 0.761). However, compared to the type strain DSMZ 44484T, the soil strains had higher SNP numbers, and wider SNP ranges per strain, than did the clinical strains (212 vs. 77 and 1–109 vs. 0–38). The gyrB-based phylogenetic relationships among the soil and human N. cyriacigeorgica strains, the three Venezuelan clinical strains, the type strain DSM 44484T, and the genome reference strain GUH-2 are shown in Figure3 and Supplementary Figure S2. The main cluster (cluster I) includes the 30 Spanish clinical strains, five soil strains, the three Venezuelan clinical strains, plus the two reference strains. Two subclusters with 20 and 17 strains were also seen, with N. cyriacigeorgica GUH-2 in one and DSM 44484T in the other. Nineteen of the 29 N. cyriacigeorgica soil strains gathered into a soil-only cluster (cluster II), i.e., it contained no human source strains. This cluster showed similarity values ranging from 91.2–92.3% with respect to the type strain. Finally, three soil strains and one clinical strain grouped into a minor cluster with two independent branches (one strain each one). The soil strains showed low resistance (0–5%) to ceftriaxone, cefepime, imipenem, amikacin, tobramycin, co-trimoxazole, and linezolid, intermediate resistance to minocycline (24%), ciprofloxacin (28%), and amoxicillin-clavulanic acid (48%), and strong resistance to clarithromycin (96.5%) (Table2). Their susceptibilities to aminoglycosides, doxycycline, tigecycline, co-trimoxazole, and linezolid were similar to those shown by the human strains. However, differences (p 0.05) were seen between the ≤ soil and clinical strains for all studied beta-lactams, fluoroquinolones, clarithromycin, and minocycline. Overall, the human N. cyriacigeorgica strains were more resistant (except for clarithromycin) than the soil strains. With respect to tigecycline (for which there are no available breakpoints for Nocardia), only one soil and three human strains returned MIC values of 4 mg/L. ≥ Microorganisms 2020, 8, 900 11 of 22

Microorganisms 2020, 8, x FOR PEER REVIEW 11 of 22

Figure 3. Phylogenetic relationships of the 29 Venezuelan N. cyriacigeorgica soil strains (in blue), three Venezuelan and 30 Spanish N. cyriacigeorgica clinical strains (in red), as revealed by their gyrB genes. The reliability of the NJ topologies was assessed by the bootstrap method (1000 replications). The asterisk indicates the strains selected for WGS. Microorganisms 2020, 8, 900 12 of 22

3.5. Whole-Genome Sequencing of the soil N. cyriacigeorgica Strains Whole-genome sequences of six soil N. cyriacigeorgica strains were obtained: two belonging to the major gyrB cluster (cluster I) and four to cluster II (the soil-only cluster). Their ANI and AAI and in silico GGDH (DDH-estimate) values, genomic G + C percentages, and other characteristics were used to determine their species. The same was performed for other N. cyriacigeorgica strains for which genomes were available (Table3). Although all the strains showed 16S rRNA identities of 99.6% ≥ with respect to the type strain and the genome of the reference strain [20], the ANI-AAI values for the GUH-2 strain were <95% [52] (except for the ANI of strain 20110626). Strain 20110626, together with 20110624 (both in cluster I), showed higher ANI-AAIs (>89.84% and >91.12% respectively) than the other four studied genomes. Determining the DDH-estimate and G + C content via the GGDH server (https://ggdc.dsmz.de/ggdc.php#)[35] showed the four selected strains (20110629, 20110639, 20110648, and 20110649) of the soil-only cluster (cluster II) did not meet the conditions of 70% DDH-estimate ≥ plus a difference of <1% G + C with respect to strains GUH-2 and DSM 44484T; they were therefore interpreted as being distinct species . In addition, lower gyrB identity ( 93.5%) and G + G content 0 0 ≤ (all 67.2%) values were seen for all the strains of the soil-only cluster than for the strains of cluster I ( 95.3% and 68.3% respectively). In contrast, for two strains of cluster I, and for those with available ≥ ≥ genomes (strains 3012STDY6756504, EML 446, EML 1456, MDA3349, MDA3732) [36], one or more criteria were met, rendering their interpretation as either “distinct or belonging to the same species”, i.e., they could not be clearly identified. To check these interpretations, analyses were run using the TrueBacTM IDBETA, AAI-profiler, and TSGY [37–39] web servers (Supplementary Table S2). Using the TrueBacTM server, the ANI values for three of the four sequenced genomes from the soil-only cluster with respect to the GUH-2 reference genome was 87.7% (0.877). With the AAI-profiler, the AAIs of the four selected strains from the soil-only cluster, and that of the EML 1456 strain, were ~75%; the remainder were over 80%. When the ≥ TYGS server (https://tygs.dsmz.de)[39] was used to determine AAI with respect to GUH-2 and DSM 44484T, the strains of the soil-only cluster returned > 1% difference in the G + C content; no such result was returned for any other strain. When these four selected strains were compared among themselves, the gyrB, ANI, AAI, DDH, and G + C ranges were 97.8–98.8%, 99.65–99.73%, 99.56–99.73%, 97.70–98.80%, and 0.01–0.33, respectively. Also using the TYGS, 16S rRNA gene sequence-based and whole-genome sequence-based trees were constructed with the above-mentioned genome sequence data and those of Nocardia type strains of other species. The soil-only cluster appeared separated from the other N. cyriacigeorgica strains in the whole-genome sequence-based tree (Figure4). Microorganisms 2020, 8, 900 13 of 22

Table 3. Comparison of the whole-genome sequences of the N. cyriacigeorgica soil strains and other stated strains, with respect to the reference genome of N. cyriacigeorgica GUH-2 (NC_016887.1) and the genome of the type strain N. cyriacigeorgica DSM 44484T.

G + C% Length DDH-Estimate (GLM-based) 16S rRNA gyrB ANI AAI Strain (ID/refSeq) (no. of Contigs; ( 70% DDH-Estimate, Difference in <1% G + C) 1,2,5 and ( 99.6%) 1,2 ( 93.5%) 1,2 ( 95%) 1,2,3 ( 95%??) 1,2,4 ≥ Depth Coverage) ≥ ≥ ≥ ≥ Interpretation DSM DSM DSM DSM Strain for Comparison GUH-2 GUH-2 GUH-2 GUH-2 GUH-2 DSM 44484T 44484T 44484T 44484T 44484T GUH-2 68.37% 6,194,645 99.9% (99.8–100%) 0.03 (either – 100 – 94.77 – 90.14 – 92.08 – NC_016887 (1) distinct or same species) DSM 44484T 68.19% 6,311,306 99.9% (99.8–100%) 0.03 (either 100 – 94.77 – 90.14 – 92.08 - – NZ_VBUR00000000.1 (1, 484x) distinct or same species) Soil Strains 20110624 68.39% 6,326,508 38.0% (36.3–41.3%) 0.05 (either 43.6% (41.1–46.1%) 0.10 (either 100 100 95.32 95.98 89.84 91.12 91.12 93.11 JAAGVC000000000 (113;79x) distinct or same species) distinct or same species) 20110626 68.29% 6,578,812 86.0% (83.4–88.3%) 0.1 (either 41.2% (38.7–3.8%) 0.04 (either 99.79 99.87 99.45 95.34 97.84 90.64 91.74 92.03 JAAGVB000000000 (158;100x) distinct or same species) distinct or same species) 20110629 66.87% 6,251,294 31.5% (29.1–34%) 1.75 31.4% (29–33.9%) 1.61 99.79 99.87 93.12 92.37 86.77 86.38 88.21 88.23 JAAGVA000000000 (71;154x) (distinct species) (distinct species) 20110639 66.95% 6,200,016 31.6% (29–34.1%), 1.43 31.5% (29.1–34%) 1.29 99.79 99.87 93.12 92.6 86.57 85.85 88.22 88.30 JAAGUZ000000000 (178;45x) (distinct species) (distinct species) 20110648 66.96% 6,274,061 31.6% (29.2–34.1%) 1.42 31.5% (29.1–34%) 1.28 99.79 99.87 92.85 92.85 86.62 85.68 88.21 88.21 JAAGUY000000000 (57;126x) (distinct species) (distinct species) 20110649 66.92% 6,258,095 31.6% (29.2–34.1%) 1.46 31.5% (29.1–34%) 1.31 99.79 99.87 92.16 91.20 86.72 85.81 88.21 88.22 JAAGUX000000000 (135;46x) (distinct species) (distinct species) Strains with available genome 3012STDY6756504 68.20% 6,476,621 39.9% (37.4–42.4%) 0.13 (either 68.8% (65.8–71.6%) 0.04 (either 100 100 96.84 97.93 89.96 96.26 92.26 96.82 NZ_LR215973.1 (1535;100x) distinct or same species) distinct or same species) EML 446 68.20% 6,520,205 41.1% (38.6–43.6%) 0.14 (either 47.0% (44.4–49.6%) 0.03 (either 100 100 97.08 97.03 90.34 91.99 92.42 93.72 NZ_VBUT00000000.1 (14;463x) distinct or same species) distinct or same species) EML 1456 68.00% 6,830,276 40.9% (38.4–43.4%) 0.34 (either 47.2% (44.6–9.8%) 0.17 (either 100 100 96.94 96.95 90.26 92.07 92.37 93.73 NZ_VBUU00000000.1 (108;458x) distinct or same species) distinct or same species) MDA3349 68.30% 6,462,637 41.3% (38.8–3.9%) 0.15 (either 81.2% (78.3–83.8%) 0.09 (either 100 100 96.55 99.84 90.21 97.83 92.30 98.00 NZ_CP026746.1 (9; 43x) distinct or same species) distinct or same species) MDA3732 68.29% 6,592,249 39.9% (37.4–42.4%) 0.13 (either 80.3% (77.4–83%) 0.03 (either 100 100 94.32 96.97 90.56 97.62 92.46 97.87 NZ_PSZF00000000.1 (84;172x) distinct or same species) distinct or same species) 1 The reference breakpoints for assigning membership to a specific species for 16S rRNA, gyrB, average nucleotide identity (ANI), average amino acid identity (AAI), in silico genome-to-genome distance similarity (GGDH; DDH-estimate) and a difference in G + C content, are indicated in brackets in the column headings. 2 Values lower than the reference breakpoints, suggestive of a distinct species, are indicated in italics. 3 ANI and coverage (range 50.94-63.08) were determined using the EzBioCloud platform (https: //www.ezbiocloud.net/tools/ani). 4 AAI at the Kostas Laboratory (http://enve-omics.ce.gatech.edu/aai). 5 DDH-estimate and difference in genomic G + C content using the DSMZ platform (https://ggdc.dsmz.de/ggdc.php#). Microorganisms 2020, 8, 900 14 of 22

Microorganisms 2020, 8, x FOR PEER REVIEW 15 of 22

Figure 4. 16S rRNA gene sequence-based and whole-genome sequence-based phylogenetic trees constructed using FastME v.2.1.6.1 software (which calculates Figure 4. 16S rRNA gene sequence‐based and whole‐genome sequence‐based phylogenetic trees constructed using FastME v.2.1.6.1 software (which Genome BLAST Distance Phylogeny (GBDP) distances; the branch lengths are scaled in terms of GBDP distance formula). The numbers above the branches are GBDP calculates Genome BLAST Distance Phylogeny (GBDP) distances; the branch lengths are scaled in terms of GBDP distance formula). The numbers above pseudo-bootstrap support values (all are >60% from 100 replications), with average branch support of 91.8% and 58.8% for the 16S rRNA gene and for the genome the branches are GBDP pseudo‐bootstrap support values (all are >60% from 100 replications), with average branch support of 91.8% and 58.8% for the 16S respectively. The trees were rooted at the midpoint. The results were provided by the Type Strain Genome Server (TYGS), a free bioinformatics platform available at rRNA gene and for the genome respectively. The trees were rooted at the midpoint. The results were provided by the Type Strain Genome Server (TYGS), a https://tygs.dsmz.de (The whole genome-based taxonomic analysis was performed on 8th January 2020) [39]. free bioinformatics platform available at https://tygs.dsmz.de (The whole genome‐based taxonomic analysis was performed on 8th January 2020) [39]. Microorganisms 2020, 8, 900 15 of 22

Using the chewBBACA platform, a novel cgMLST typing method based on 3048 loci was performed independent of any defined comparator strain [40]. The N. cyriacigeorgica type strain designated IMMIB D-1627 has several culture collection denominations, including DSM 44484 and NBRC 100375 (although their respective genomes differ in 10 alleles). In the cgMLST dendrogram, the genome of the type strain NBRC 100375 has a central node from which other genomes emerge. Indeed, moving in a clockwise fashion, six distinct lineages can be seen (Figure5), with the genome of the reference GUH-2 appearing as lineage 1 (with 3034 different alleles). The genomes of the soil strains appeared as lineages 1, 3, and 5, with lineage 5 belonging to the soil-only cluster. The strains of this latter cluster differ in 3047 alleles with respect to the central node of NBRC 100375, and among themselves by a mean 1594 alleles. Using the 107 essential single-copy genes extracted by BCGtree analysis [41], the four selected strains of the soil-only cluster grouped into one of two clusters with a high bootstrap value (Figure5). By the use of different platforms [43,44,47,48], the ast-1 beta-lactamase gene (class A beta-lactamase) was detected in strain with decreased amoxicillin-clavulanate acid susceptibility, CNM20110626, which showed a 98.06% of identity (new allele with 6 amino acids changes) respective to its counterpart in N. cyriacigeorgica GUH-2 strain. As well as vanRS, the two-component system response regulator of the glycopeptide resistance gene cluster, in the NCBR 100375, 3012STDY6756504, MDA3349, EML446, MDA3732 and EML1456 strains. With KOALA for KEGG Orthology [46], more putative antimicrobial genes were identified in both groups of strains as some aminoglycoside resistance genes (strB, streptomycin 6-kinase; aadA, streptomycin 3”-adenylyltransferase; and aph3-II, aminoglycoside 30-phosphotransferase II), macrolide resistance genes (ermC/A, 23S rRNA methyltransferases; carA, transport system ATP-binding/permease protein; ereA_B, erythromycin esterase; vat, virginiamycin A acetyltransferase; vgb, virginiamycin B lyase), chloramphenicol resistance protein (cmlR, MFS transporter), vancomycin resistance (vanY, zinc D-Ala-D-Ala carboxypeptidase), and multidrug resistance efflux pumps genes of MexJK-OprM, MexPQ-OpmE, and QacA). The chloramphenicol 3-O phosphotransferase cpt, was detected in GUH-2 and 3012STDY6756504 strains. To note, the aadA gene was detected in strains of soil-only cluster strains together with strB, but not with aph3-II. The remaining strains only have aph3-II with strB, except CNM20110624 and 3012STDY6756504 strains with aadA/aph3-II/ strB genes. Regarding the quinolone resistance, the topoisomerase subunits GyrA/B of the soil N. cyriacigeorgica strains showed both two major alleles (19 and 11 changes, respectively): GyrA1 and GyrB1, in CNM20110624-626 (with 6 and 4 differences outside of the quinolone-determining-region between both strains); and GyrA2 and GyrB2, without changes in the four strains of the soil-only cluster (Supplementary Figure S3). Lastly, no intact or questionable phages were detected in the soil strains. Microorganisms 2020, 8, 900 16 of 22 Microorganisms 2020, 8, x FOR PEER REVIEW 16 of 22

Figure 5. Left: Phylogenetic tree constructed by MAFFT alignment and neighbor-joining with the Clustal W2 algorithm, based on the cgMLST associations among the . N. cyriacigeorgica genomes. The tree was built using chewBBACA software and based on 3048 loci. [40]. The DSM 44484T and NBRC 100375 genomes correspond to the N. cyriacigeorgica type strain IMMIB D-1627; GUH-2 is the reference genome. Branches indicate the number of diﬀerent alleles. Right: maximum likelihood Figure 5. Left: Phylogenetic tree constructed by MAFFT alignment and neighbor‐joining with the Clustal W2 algorithm, based on the cgMLST associations among phylogenetic tree produced with a concatenated gene matrix with 107 conserved proteins using RAxML v. 8.2.9 and bcgTree software v.1.1.0 software (100 bootstrap the N. cyriacigeorgica genomes. The tree was built using chewBBACA software and based on 3048 loci. [40]. The DSM 44484T and NBRC 100375 genomes correspond replications) [41]. The N. cyriacigeorgica soil strains are colored blue and the Nocardia clinical strains red. The percentage of bootstrap replicate trees (1000 replications) to the N. cyriacigeorgica type strain IMMIB D‐1627; GUH‐2 is the reference genome. Branches indicate the number of different alleles. Right: maximum likelihood in which the associated taxa clustered together are shown next to the branches. Bar: 0.02 changes per nucleotide position. phylogenetic tree produced with a concatenated gene matrix with 107 conserved proteins using RAxML v. 8.2.9 and bcgTree software v.1.1.0 software (100 bootstrap replications) [41]. The N. cyriacigeorgica soil strains are colored blue and the Nocardia clinical strains red. The percentage of bootstrap replicate trees (1000 replications) in which the associated taxa clustered together are shown next to the branches. Bar: 0.02 changes per nucleotide position. Microorganisms 2020, 8, 900 17 of 22

4. Discussion Like other actinomycetes, Nocardia spp. contribute to soil health, playing major roles in the cycling of organic matter, inhibiting the growth of plant pathogens, and decomposing complex mixtures of dead plants and animals [1,53]. As well as maintaining the biotic equilibrium of the soil, these bacteria are involved in a wide array of opportunistic infections in both immunocompromised and immunocompetent persons [13]. Mycetoma and pulmonary nocardiosis, respectively caused by traumatic inoculation and inhalation, are the most common [14,15]. The increase in the size of the immunocompromised and immunosenescent populations has led to an increase in the number of cases of nocardiosis recorded. The annual incidence rate in Canada has now reached 0.87/100,000 inhabitants [15]; in Western Europe, the hospitalization rate due to nocardiosis has reached 0.04/100,000 inhabitants [54]. Climate, vegetation type, and soil pH probably affect the frequency and diversity of soil aerobic Actinobacteria [55]. Those that cause human infections in any given area are typically those found in the local soil [56]. Thus, different Nocardia species appear as major aetiological agents in different countries. For instance, N. farcinica causes infections in Canada [15], France [57], and Japan [58], but not in Spain, where the incidence N. cyriacigeorgica is double that of N. farcinica [59]. Nocardia spp. in the environment, thus posing some risk to human health [14], a fact reflected in the greater incidence of mycetoma in farmers and other people from rural areas of Lara State [16]. Nocardia genus contains about 200 species (https://lpsn.dsmz.de/), however, in the present work, only seven species were identified, with N. cyriacigeorgica the most common (71.8%). Surprisingly, Nocardia brasiliensis, the main causal agent of mycetoma in Lara State, was not isolated in the previous work [16]. In south-eastern Spain, N. cyriacigeorgica (previously identified as the N. asteroides complex) [60] has been detected in soil samples [61], and it is responsible for the majority of human nocardiosis (25%) [59]. However, N. brasiliensis, which is responsible for more than half of soft tissue/bone infections [59], was not detected in the above study [61]. This might be explained in that actinomycetes are 3–5.6 times more abundant in air samples above ground than in the soil [62]. With respect to the present molecular targets, MLSA (gyrB-16S rRNA-secA1-hsp65) was the arbiter of Nocardia species identification [25], confirming the 16S rRNA- and gyrB-based assignment results for 89.47% and 71.05% of the strains, respectively. Nearly 70% of the soil N. cyriacigeorgica strains isolated from five of the nine sampling sites gathered into MLSA cluster B or gyrB cluster II (the soil-only cluster). Both clustering methods are valuable in species/subspecies identification [63], although the gyrB method, with just one studied gene, is more simple. gyrB gene sequencing showed the soil-derived N. cyriacigeorgica strains to be less diverse (lower HGDI) than the human-isolated strains, although the number and range of SNPs per strain were significantly greater. The difference in SNPs found between the DSM 44484T strain and the soil strains might suggest the presence of some atypical N. cyriacigeorgica strains. In addition, the soil strains were more susceptible to beta-lactams, fluoroquinolones, and minocycline than were the human strains, and more resistant to clarithromycin. Regarding fluoroquinolones, susceptibility differences could be related to variations in the amino acid composition of GyrA/GyrB. These differences might be the result of reduced exposure to antimicrobials in Venezuelan soils, or perhaps low intrinsic resistance of this variant. To check the species assignment of the strains in the soil-only cluster - despite them belonging to the same species according to their 16S rRNA results - some were subjected to WGS along with others from outside this cluster. Several coefficients were required to reach specific thresholds for an assignment to be deemed correct: > 95% ANI/AAI, > 70% DDH-estimate and a < 1% difference in G + C content [34,52,64,65]. ANI resolves well between genomes that share 80–100% identity, and AAI does so for species that share < 80% ANI and/or when 30% of their gene content is very divergent [34]. In the present work, the results of both AAI and ANI were taken into account, along with the DDH-estimate, and the G + C content since a query genome with an ANI of < 95% likely represents a new species [66]. Indeed, with respect to the genome of reference strain GUH-2, average ANI and AAI values of around 87% were returned for the strains of the soil-only cluster, along with a mean DDH-estimate of 31.6% Microorganisms 2020, 8, 900 18 of 22 and G + C content differences of around 1.5%. In addition, the N. cyriacigeorgica strains of the soil-only cluster showed the greatest identity among themselves, with average ANI values of 99.7% being returned. For two strains from outside of the soil-only cluster, as well as for those for which genomes were available, the ANI and AAI values were around 90% and 92%, satisfying the criterion of a <1% difference in the G + C content, meaning they belong to the same species. According to the commercial TrueBacTM system, and the AAI-profiler and TSGY systems (both open source) [37–39], the genomic evidence might suggest that a new species (Nocardia venezuelensis sp. nov) exists among the soil-only cluster strains examined, all of which had low G + C contents. Some of the available genomes studied might also belong to a new species. In cgMLST (performed using chewBBACA software) [40], six lineages appear for 12 N. cyriacigeorgica strains around the reference strain DSM 44484T/ NRBC 100375. It may be that the DSM 44484T strain provides a better reference genome than the current GUH-2 reference strain. Intraspecies MLSA sub-clusters of N. cyriacigeorgica have already been described [63]; thus, whole-genome analyses of N. cyriacigeorgica should be performed to determine its lineages, with the description of different species/subspecies as members of a single complex. In conclusion, no genetic differences, nor differences in antimicrobial susceptibilities, were found between the Nocardia strains isolated from the Venezuelan soil samples and the reference or clinical strains–except for the strains of N. cyriacigeorgica-. This might indicate that some of the latter belong to a new subspecies of N. cyriacigeorgica or even a new species. Should this be confirmed, the name Nocardia venezuelensis is proposed.

Supplementary Materials: The following are available online at http://www.mdpi.com/2076-2607/8/6/900/s1, Figure S1. Phylogenetic ML tree based on the MLSA analysis gyrB-16S rRNA-secA-hsp65 genes) of the 38 N. cyriacigeorgica strains from soil (in blue), 5 Nocardia clinical strains from Venezuelan patients and nine Spanish clinical strains representing each species present in soil (in red), plus the type strains (in black). The asterisk indicates the strains selected for WGS. Na stands for N. abscessus, so on Nast for N. mexicana, Nc for N. cyriacigeorgica, Ne for N. elegans, Nf for N. farcinica, Ni for N. ignorata, Nm for N. mexicana, Nn for N. nova, Nr for N. rhamnosiphila, and Nv for N. vermiculata. The reliability of the topologies was assessed by the bootstrap method with 1000 replicates; Figure S2. Phylogenetic relationships of the 29 Venezuelan N. cyriacigeorgica soil strains (in blue), three Venezuelan and 30 Spanish N. cyriacigeorgica clinical strains (in red), as revealed by their gyrB genes. The reliability of the ML topologies was assessed by the bootstrap method (1000 replications). The asterisk indicates the strains selected for WGS; Figure S3. Amino acid sequences of GyrA alignment from N. cyriacigeorgica genome reference strain GUH-2, type strain DSM44484T, CNM20110626, and CNM20110624 with major allele GyrA1, and the strains of the soil-only cluster (CNM20110629, CNM20110639, CNM20110648, and CNM20110649) with major allele GyrA2; Table S1. Comparison of the main typing characteristics (16S rDNA, gyrB, and GyrB) between the N. cyriacigeorgica soil strains from Lara State (Venezuela) and clinical samples from Spanish patients; Table S2. Interpretations of the analysis of the genomes of the soil N. cyriacigeorgica strains and the NCBI-available N. cyriacigeorgica genomes in terms of gyrB, ANI, AAI, in silico genome-to-genome distance similarity (GGDH; DDH-estimate), and differences in G+C content. Author Contributions: G.C., M.S.S. and S.V. contributed conception and design of the study; E.G. and A.R. collected the soil strains; G.C., M.S.S., E.G., M.J.M.-P., and P.V. performed the molecular studies and participated in the data analysis of soil and clinical strains; S.V. done the susceptibility testing; N.G. and P.J. carried out the whole-genome sequencing; S.M. and I.C. conducted the bioinformatic analysis; G.C., S.M. and S.V. wrote the first draft of the manuscript. All authors contributed to manuscript revision, read and approved the submitted version. All authors have read and agreed to the published version of the manuscript. Funding: This work was funded by a grant to N.G. from the Instituto de Salud Carlos III (MPY 1278/15). A part of this manuscript forms the M.S project undertaken at the Faculty of Science, Universidad Autónoma de Madrid (Spain). Acknowledgments: The authors are grateful to M. Blanco (Universidad de Los Andes, Mérida, Venezuela) for assistance in strain isolation, to J. A. Saéz-Nieto and A. Zaballos for help in the identification and sequencing, to Adrian Burton for language and editing assistance, and to the health laboratories that provided the clinical Nocardia strains examined. Conflicts of Interest: The authors declare no conflicts of interest. Microorganisms 2020, 8, 900 19 of 22

References

1. Goodfellow, M. “Family IV. Nocardiaceae”. In Bergey’s Manual of Systematic Bacteriology; Whitman, W., Goodfellow, M., Kämpfer, P., Busse, H.J., Trujillo, M., Ludwig, W., Suzuki, K., Eds.; Springer: New York, NY, USA, 2012. 2. Mohammadipanah, F.; Wink, J. Actinobacteria from Arid and Desert Habitats: Diversity and Biological Activity. Front Microbiol. 2016, 6, 1541. [CrossRef][PubMed] 3. Yaemsiri, S.; Sykes, J.E. Successful Treatment of Disseminated Nocardiosis Caused by Nocardia veterana in a Dog. J. Vet. Intern Med. 2018, 32, 418–422. [CrossRef][PubMed] 4. Chen, J.; Tan, W.; Wang, W.; Hou, S.; Chen, G.; Xia, L.; Lu, Y. Identification of common antigens of three pathogenic Nocardia species and development of DNA vaccine against fish nocardiosis. Fish Shellfish Immunol. 2019, 95, 357–367. [CrossRef][PubMed] 5. Yasuike, M.; Nishiki, I.; Iwasaki, Y.; Nakamura, Y.; Fujiwara, A.; Shimahara, Y.; Kamaishi, T.; Yoshida, T.; Nagai, S.; Kobayashi, T.; et al. Analysis of the complete genome sequence of Nocardia seriolae UTF1, the causative agent of fish nocardiosis: The first reference genome sequence of the fish pathogenic Nocardia species. PLoS ONE 2017, 3, 12. [CrossRef] 6. Bull, A.T.; Asenjo, J.A. Microbiology of hyper-arid environments: Recent insights from the Atacama Desert, Chile. Van Leeuw J. Microb. 2013, 103, 1173–1179. [CrossRef] 7. Luo, Q.; Hiessl, S.; Steinbüchel, A. Functional diversity of Nocardia in metabolism. Environ. Microbiol. 2014, 16, 29–48. [CrossRef] 8. Sharma, P.; Kalita, M.C.; Thakur, D. Broad Spectrum Antimicrobial Activity of Forest-Derived Soil Actinomycete, Nocardia sp. PB-52. Front Microbiol. 2016, 7, 347. [CrossRef] 9. Shivlata, L.; Satyanarayana, T. Thermophilic and alkaliphilic Actinobacteria: Biology and potential applications. Front Microbiol. 2015, 6, 1014. [CrossRef] 10. Dhakal, D.; Pokhrel, A.R.; Shrestha, B.; Sohng, J.K. Marine Rare Actinobacteria: Isolation, Characterization, and Strategies for Harnessing Bioactive Compounds. Front Microbiol. 2017, 8, 1106. [CrossRef] 11. Lara-Severino, R.D.; Camacho-López, M.A.; Casanova-González, E.; Gómez-Oliván, L.M.; Sandoval-Trujillo, Á.H.; Isaac-Olivé, K.; Ramírez-Durán, N. Haloalkalitolerant Actinobacteria with capacity for anthracene degradation isolated from soils close to areas with oil activity in the State of Veracruz, Mexico. Int. Microbiol. 2016, 19, 15–26. [CrossRef] 12. Rodrigues, E.M.; Vidigal, P.M.P.; Pylro, V.S.; Morais, D.K.; Leite, L.R.; Roesch, L.F.W.; Tótola, M.R. Draft genome of Nocardia farcinica TRH1, a linear and polycyclic aromatic hydrocarbon-degrading bacterium isolated from the coast of Trindade Island, Brazil. Braz. J. Microbiol. 2017, 48, 391–392. [CrossRef][PubMed] 13. Brown-Elliott, B.A.; Brown, J.M.; Conville, P.S.; Wallace, R.J. Clinical and laboratory features of the Nocardia spp. based on current molecular taxonomy. Clin. Microbiol. Rev. 2006, 19, 259–282. [CrossRef][PubMed] 14. Ambrosioni, J.; Lew, D.; Garbino, J. Nocardiosis: Updated clinical review and experience at a tertiary center. Infection 2010, 38, 89–97. [CrossRef][PubMed] 15. Tremblay, J.; Thibert, L.; Alarie, I.; Valiquette, L.; Pépin, J. Nocardiosis in Quebec, Canada, 1988–2008. Clin. Microbiol. Infect. 2011, 17, 690–696. [CrossRef] 16. Ramírez, A.; Blanco, M.; García, E. Biogeografía de Nocardia: Estudio de la población edáfica de Nocardia en diversas zonas climáticas del Estado de Lara. Rev. Soc. Ven. Microbiol. 2003, 23, 1–7. 17. Carrasco, G.; Valdezate, S.; Garrido, N.; Villalón, P.; Medina-Pascual, M.J.; Sáez-Nieto, J.A. Identification, typing, and phylogenetic relationships of the main clinical Nocardia species in Spain according to their gyrB and rpoB genes. J. Clin. Microbiol. 2013, 51, 3602–3608. [CrossRef] 18. Drancourt, M.; Bollet, C.; Carlioz, A.; Martelin, R.; Gayral, J.P.; Raoult, D. 16S ribosomal DNA sequence analysis of a large collection of environmental and clinical unidentifiable bacterial isolates. J. Clin. Microbiol. 2000, 38, 3623–3630. [CrossRef] 19. Takeda, K.; Kang, Y.; Yazawa, K.; Gonoi, T.; Mikami, Y. Phylogenetic studies of Nocardia species based on gyrB gene analyses. J. Med. Microbiol. 2010, 59, 165–171. [CrossRef] 20. Clinical and Laboratory Standards Institute. Susceptibility Testing of Mycobacteria, Nocardia spp., and Other Aerobic Actinomycetes, 3rd ed.; Clinical and Laboratory Standards Institute: Wayne, PA, USA, 2018. 21. Hall, T.A. BioEdit: A user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp Ser. 1999, 41, 95–98. Microorganisms 2020, 8, 900 20 of 22

22. Librado, P.; Rozas, J. DnaSP v5: A software for comprehensive analysis of DNA polymorphism data. Bioinformatics 2009, 25, 1451–1452. [CrossRef] 23. Tamura, K.; Stecher, G.; Peterson, D.; Filipski, A.; Kumar, S. MEGA6: Molecular Evolutionary Genetics Analysis version 6.0. Mol. Biol. Evol. 2013, 30, 2725–2729. [CrossRef][PubMed] 24. Gascuel, O. BIONJ: An improved version of the NJ algorithm based on a simple model of sequence data. Mol. Biol. Evol. 1997, 14, 685–695. [CrossRef][PubMed] 25. McTaggart, L.R.; Richardson, S.E.; Witkowska, M.; Zhang, S.X. Phylogeny and identification of Nocardia species on the basis of multilocus sequence analysis. J. Clin. Microbiol. 2010, 48, 4525–4533. [CrossRef] [PubMed] 26. Hunter, P.R.; Gaston, M.A. Numerical index of the discriminatory ability of typing systems: An application of Simpson’s index of diversity. J. Clin. Microbiol. 1988, 26, 2465–2466. [CrossRef] 27. Clinical Laboratory Standards Institute [CLSI]. Susceptibility Testing of Mycobacteria, Nocardiae, and Other Aerobic Actinomycetes. Approved Standard-M24-A2, 2nd ed.; Clinical and Laboratory Standards Institute: Wayne, PA, USA, 2011. 28. Bolger, A.M.; Lohse, M.; Usadel, B. Trimmomatic: A flexible trimmer for Illumina Sequence Data. Bioinformatics 2014, 30, 2114–2120. [CrossRef] 29. Larsen, M.V.; Cosentino, S.; Lukjancenko, O.; Saputra, D.; Rasmussen, S.; Hasman, H.; Sicheritz-Pontén, T.; Aarestrup, F.M.; Ussery, D.W.; Lund, O. Benchmarking of methods for genomic taxonomy. J. Clin. Microbiol. 2014, 52, 1529–1539. [CrossRef] 30. Bankevich, A.; Nurk, S.; Antipov, D.; Gurevich, A.A.; Dvorkin, M.; Kulikov, A.S.; Lesin, V.M.; Nikolenko, S.I.; Pham, S.; Prjibelski, A.D.; et al. SPAdes: A new genome assembly algorithm and its applications to single-cell sequencing. J. Comput. Biol. 2012, 19, 455–477. [CrossRef] 31. Seemann, T. Prokka: Rapid prokaryotic genome annotation. Bioinformatics 2014, 30, 2068–2069. [CrossRef] 32. Gurevich, A.; Saveliev, V.; Vyahhi, N.; Tesler, G. QUAST: Quality assessment tool for genome assemblies. Bioinformatics 2013, 29, 1072–1075. [CrossRef] 33. Yoon, S.H.; Ha, S.M.; Lim, J.M.; Kwon, S.J.; Chun, J. A large-scale evaluation of algorithms to calculate average nucleotide identity. Ant. V. Leeuw. 2017, 110, 1281–1286. [CrossRef] 34. Rodriguez, R.L.M.; Konstantinidis, K.T. The enveomics collection: A toolbox for specialized analyses of microbial genomes and metagenomes (No. e1900v1). PeerJ Preprints 2016. 35. Meier-Kolthoff, J.P.; Auch, A.F.; Klenk, H.P.; Göker, M. Genome sequence-based species delimitation with confidence intervals and improved distance functions. BMC Bioinf. 2013, 14, 60. [CrossRef] 36. Vautrin, F.; Bergeron, E.; Dubost, A.; Abrouk, D.; Martin, C.; Cournoyer, B.; Louzier, V.; Winiarski, T.; Rodriguez-Nava, V.; Pujic, P. Genome Sequences of Three Nocardia cyriacigeorgica Strains and One Nocardia asteroides Strain. Microbiol. Resour. Announc. 2019, 15, 8. [CrossRef][PubMed] 37. Medlar, A.J.; Törönen, P.; Holm, L. AAI-profiler: Fast proteome-wide exploratory analysis reveals taxonomic identity, misclassification and contamination. Nucleic Acids Res. 2018, 46, 479–485. [CrossRef][PubMed] 38. Ha, S.M.; Kim, C.K.; Roh, J.; Byun, J.H.; Yang, S.J.; Choi, S.B.; Chun, J.; Yong, D. Application of the Whole Genome-Based Bacterial Identification System, TrueBac ID, Using Clinical Isolates That Were Not Identified With Three Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry (MALDI-TOF MS) Systems. Ann. Lab. Med. 2019, 39, 530–536. [CrossRef][PubMed] 39. Meier-Kolthoff, J.P.; Göker, M. TYGS is an automated high-throughput platform for state-of-the-art genome-based taxonomy. Nat. Commun. 2019, 10, 2182. [CrossRef] 40. Silva, M.; Machado, M.P.; Silva, D.N.; Rossi, M.; Moran-Gilad, J.; Santos, S.; Ramirez, M.; Carriço, J.A. chewBBACA: A complete suite for gene-by-gene schema creation and strain identification. Microb. Genom. 2018, 4, 3. [CrossRef][PubMed] 41. Ankenbrand, M.J.; Keller, A. bcgTree: Automatized phylogenetic tree building from bacterial core genomes. Genome 2016, 59, 783–791. [CrossRef][PubMed] 42. Stamatakis, A. RAxML version 8: A tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics 2014, 30, 1312–1313. [CrossRef][PubMed] 43. Zankari, E.; Hasman, H.; Cosentino, S.; Vestergaard, M.; Rasmussen, S.; Lund, O.; Aarestrup, F.M.; Larsen, M.V. Identification of acquired antimicrobial resistance genes. J. Antimicrob. Chemother. 2012, 67, 2640–2644. [CrossRef][PubMed] Microorganisms 2020, 8, 900 21 of 22

44. Alanjary,M.; Kronmiller, B.; Adamek, M.; Blin, K.; Weber, T.; Huson, D.; Philmus, B.; Ziemert, N. The Antibiotic Resistant Target Seeker (ARTS), an exploration engine for antibiotic cluster prioritization and novel drug target discovery. Nucleic Acids Res. 2017, 45, 42–48. [CrossRef][PubMed] 45. Guitor, A.K.; Raphenya, A.R.; Klunk, J.; Kuch, M.; Alcock, B.; Surette, M.G.; McArthur, A.G.; Poinar, H.N.; Wright, G.D. Capturing the Resistome: A Targeted Capture Method To Reveal Antibiotic Resistance Determinants in Metagenomes. Antimicrob. Agents Chemother. 2019, 20, 64. [CrossRef][PubMed] 46. Kanehisa, M.; Sato, Y.; Morishima, K. BlastKOALA and GhostKOALA: KEGG Tools for Functional Characterization of Genome and Metagenome Sequences. J. Mol. Biol. 2016, 428, 726–731. [CrossRef] [PubMed] 47. Inouye, M.; Dashnow, H.; Raven, L.A.; Schultz, M.B.; Pope, B.J.; Tomita, T.; Zolber, J.; Holt, K.E. SRST2: Rapid genomic surveillance for public health and hospital microbiology labs. Genome Med. 2014, 6, 90. [CrossRef] 48. Gupta, S.K.; Padmanabhan, B.R.; Diene, S.M.; Lopez-Rojas, R.; Kempf, M.; Landraud, L.; Rolain, J.M. ARG-ANNOT, a new bioinformatic tool to discover antibiotic resistance genes in bacterial genomes. Antimicrob. Agents Chemother. 2014, 58, 212–220. [CrossRef][PubMed] 49. Arndt, D.; Grant, J.R.; Marcu, A.; Sajed, T.; Pon, A.; Liang, Y.; Wishart, D.S. PHASTER: A better, faster version of the PHAST phage search tool. Nucleic Acids Res. 2016, 44, 16–21. [CrossRef] 50. Clinical and Laboratory Standards Institute. Interpretive Criteria for Identification of Bacteria and Fungi by DNA Target Sequencing: Approved Guideline MM18-A; CLSI: Wayne, PA, USA, 2008. 51. Brown-Elliott, B.A.; Conville, P.; Wallace, R.J. Current Status of Nocardia Taxonomy and Recommended Identification Methods. Clin. Microbiol. Newsl. 2015, 37, 25–32. [CrossRef] 52. Chun, J.; Oren, A.; Ventosa, A.; Christensen, H.; Arahal, D.R.; da Costa, M.S.; Rooney, A.P.; Yi, H.; Xu, X.W.; De Meyer, S.; et al. Proposed minimal standards for the use of genome data for the taxonomy of prokaryotes. Int. J. Syst. Evol. Microbiol. 2018, 68, 461–466. [CrossRef] 53. Bhatti, A.A.; Haq, S.; Bhat, R.A. Actinomycetes benefaction role in soil and plant health. Microb. Pathog. 2017, 111, 458–467. [CrossRef] 54. Ott, S.R.; Meier, N.; Kolditz, M.; Bauerd, T.T.; Rohde, G.; Presterl, E.; Schürmann, D.; Lepper, P.M.; Ringshausen, F.C.; Flick, H.; et al. Pulmonary nocardiosis in Western Europe-Clinical evaluation of 43 patients and population-based estimates of hospitalization rates. Int. J. Inf. Dis. 2019, 81, 140–148. [CrossRef] [PubMed] 55. Kachuei, R.; Emami, M.; Mirnejad, R.; Khoobdel, M. Diversity and frequency of Nocardia spp. in the soil of Isfahan province, Iran. Asian Pac. J. Trop. Biomed. 2012, 2, 474–478. [CrossRef] 56. Aghamirian, M.; Ghiasian, S.A. Isolation and characterization of medically important aerobic actinomycetes in soil of Iran (2006-2007). Open Microbio. L J. 2009, 3, 53–57. [CrossRef] 57. Lebeaux, D.; Bergeron, E.; Berthet, J.; Djadi-Prat, J.; Mouniée, D.; Boiron, P.; Lortholary, O.; Rodriguez-Nava, V. Antibiotic susceptibility testing and species identification of Nocardia isolates: A retrospective analysis of data from a French expert laboratory, 2010–2015. Clin. Microbiol. Infec. 2019, 25, 489–495. [CrossRef][PubMed] 58. Kageyama, A.; Yazawa, K.; Ishikawa, J.; Hotta, K.; Nishimura, K.; Mikami, Y. Nocardial infections in Japan from 1992 to 2001, including the first report of infection by Nocardia transvalensis. Eur. J. Epidemiol. 2004, 19, 383–389. [CrossRef] 59. Valdezate, S.; Garrido, N.; Carrasco, G.; Medina-Pascual, M.J.; Villalón, P.; Navarro, A.M.; Sáez-Nieto, J.A. Epidemiology and susceptibility to antimicrobial agents of the main Nocardia species in Spain. J. Antimicrob. Chemother. 2017, 72, 754–761. [CrossRef][PubMed] 60. Conville, P.S.; Brown-Elliott, B.A.; Smith, T.; Zelazny, A.M. The Complexities of Nocardia Taxonomy and Identification. J. Clin. Microbiol. 2017, 56, e01419-17. [CrossRef] 61. Valero-Guillén, P.L.; Martín-Luengo, F. Nocardia in soils of southeastern Spain: Abundance, distribution, and chemical characterisation. Can. J. Microbiol. 1984, 30, 1088–1092. [CrossRef][PubMed] 62. Weber, C.F.; Werth, J.T. Is the lower atmosphere a readily accessible reservoir of culturable, antimicrobial compound-producing Actinomycetales? Front Microbiol. 2015, 6, 802. [CrossRef][PubMed] 63. Wei, M.; Wang, P.; Yang, C.; Gu, L. Molecular identification and phylogenetic relationships of clinical Nocardia isolates. Ant. V. Leeuw. 2019, 112, 1755–1766. [CrossRef][PubMed] 64. Thompson, C.C.; Chimetto, L.; Edwards, R.A.; Swings, J.; Stackebrandt, E.; Thompson, F.L. Microbial genomic taxonomy. BMC Genom. 2013, 14, 913. [CrossRef][PubMed] Microorganisms 2020, 8, 900 22 of 22

65. Varghese, N.J.; Mukherjee, S.; Ivanova, N.; Konstantinidis, K.T.; Mavrommatis, K.; Kyrpides, N.K.; Pati, A. Microbial species delineation using whole genome sequences. Nucleic Acids Res. 2015, 43, 6761–6771. [CrossRef][PubMed] 66. Rodriguez, R.L.M.; Gunturu, S.; Harvey, W.T.; Rosselló-Mora, R.; Tiedje, J.M.; Cole, J.R.; Konstantinidis, K.T. The Microbial Genomes Atlas (MiGA) webserver: Taxonomic and gene diversity analysis of Archaea and Bacteria at the whole genome level. Nucleic Acids Res. 2018, 46, 282–288. [CrossRef][PubMed]

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).